University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln
US Department of Energy Publications
US Department of Energy
1-1-1994
Interaction of Hydrophobic Organic Compounds
with Mineral-Bound Humic Substances
Ellyn M. Murphy
Pacific Northwest National Laboratory
John M. Zachara
Pacific Northwest National Laboratory, [email protected]
Steven Smith
Pacific Northwest National Laboratory
Jerry Phillips
Pacific Northwest National Laboratory
Thomas Wletsma
Pacific Northwest National Laboratory
Follow this and additional works at: http://digitalcommons.unl.edu/usdoepub
Part of the Bioresource and Agricultural Engineering Commons
Murphy, Ellyn M.; Zachara, John M.; Smith, Steven; Phillips, Jerry; and Wletsma, Thomas, "Interaction of Hydrophobic Organic
Compounds with Mineral-Bound Humic Substances" (1994). US Department of Energy Publications. Paper 212.
http://digitalcommons.unl.edu/usdoepub/212
This Article is brought to you for free and open access by the US Department of Energy at DigitalCommons@University of Nebraska - Lincoln. It has
been accepted for inclusion in US Department of Energy Publications by an authorized administrator of DigitalCommons@University of Nebraska Lincoln.
Environ. Sci. Technol. 1994, 28, 1291-1299
Interaction of Hydrophobic Organic Compounds with Mineral-Bound Humic
Substances
Eiiyn M. Murphy,’ John M. Zachara, Steven C. Smith, Jerry L. Phillips, and Thomas W. Wletsma
Interfacial Geochemistry Group, Pacific Northwest Laboratory, Richland, Washington 99352
The sorption of hydrophobic organic comPounds
on mineral-associated peat humic acid @‘HA) was evaluated under different pH and electrolyte regimes. Relative
size distribution measurements indicated that PHA was
“coiled”in solution at high ionic strength (I)and elongated
at low I. The sorption of PHA to hematite and kaolinite
varied with I and electrolyte cation, suggesting that the
configuration of the humic acid in solution influenced its
structure on the mineral surface. The sorption maxima
for PHA ori kaolinite indicated that PHA occupies twice
the mineral surface area at low I(0.005)as that observed
at high I (o.l). HOC sorption to mineral-bound PHA in
Na+ electrolyte was greater at lower I , indicating that
humate structure was a plausible determinant of HOC
sorption. Freundlich isotherms of dibenzothiophene on
the PHA-coated kaolinite did not display unit slope,
regardless of pH, I, or cation. Carbazole and anthracene
displayed competitive behavior for sorption onto PHAcoated kaolinite. Collectively, the experimental observations indicate that hydrophobic adsorption rather than
phase partitioning was the dominant mode of HOC
binding.
Introduction
The aqueous concentrations and transport of hydrophobic organic compounds (HOC) in subsurface systems
are influenced by sorption to organic matter (1-5) and
binding to dissolved humic substances (6-1 3). Natural
dissolved/colloidal organic material binds to a variety of
aluminum and iron oxides and layer silicate surfaces that
are common in aquifers (14-22). Thus, dissolved humic
substances may form organic coatings on aquifer sediments
that may in turn facilitate the sorption of HOC.
The sorption of HOC to mineral-bound organic matter
is influenced by the chemical properties of the humic
material (23-29). The structural orientation of sorbed
humic substances may be controlled by the distribution
of hydroxyl groups on the mineral surface, and this
structure can also influence HOC sorption (2, 3). Furthermore, the surface structure of mineral-bound humic
substances may be controlled by their solution conformation, as has been observed for polyelectrolytes (30-33).
The structure of dissolved humic substances is affected
by pH, ionic strength (I),and electrolyte cation valence
(17,18,34-36). The configuration and sorption behavior
of humic substances are analogous to those of polyelectrolytes, which have received more research attention. At
high I and low pH, the charge repulsion between adjacent
carboxyl or hydroxyl groups on the humic substance is
neutralized by solvating cations, resulting in a coiled
configuration (18, 34, 35). The sorption of the “coiled”
humic substance may be similar to that of an uncharged
polymer that displays a loop and tail formation (37) with
* E-mail address:
[email protected],
0013-936X/94/0928-1291$04.50/0
few attachment points to the mineral surface (18,32,33).
At low I and high pH, the humic substance may adopt a
open configuration in solution as a result of charge
between ionized functional groups (37-39).This
points of atextended configuration may create
tachment to the mineral surface (39).
Electrolyte cation valence also affects the configuration
of humic acid in solution. Calcium increases the sorption
of humic substances to goethite, presumably by the
interaction of Ca2+ with carboxyl groups on the humic
acid (18, 37, 40, 41) and decreases the negative electrophoretic mobility of the humic oxide complex. Calcium
increases sorption and the initial slope of the isotherm of
polyacrylic acids (average MW 4000, 2000, and 700) to
Ti02 and decreases the dependence of sorption on
molecular weight (37). Calcium is believed to form a
binuclear complex with polyacrylic acid (37,42),internally
groups that affect the dimension and
bridging
rigidity of the polyacrylic acid structure (43).
Although research has focused on the effects of I , pH,
and cation Valence on the configuration Of humic substances and polyelectrolytes in aqueous solutions, it is not
known whether these same variables influence the configuration and structure of mineral-bound humic substances. Given that natural waters show a wide range in
chemical composition, it is important to establish the
influence of aqueous chemical variables on the sorptivity
of organic matter coatings. In this study, we evaluated
the effects of I , pH, and cation valence on HOC sorption
to humic-coated mineral solids to determine whether HOC
sorption trends indicate structural differencesin the sorbed
humic coating. In addition, the sorption of HOC to the
mineral-bound organic matter was evaluated for its
consistency with phase partitioning or hydrophobic adsorption.
Experimental Procedures
Materials. A well-crystallized kaolinite (KGa-1, Clay
Minerals Society E h r c e Clay) was prepared according to
the ~ - ~ ~ e t hofo dJohnston
s
et a]. (44)and Zachara et al. (45).
This involved the separation of the 0.2-2.0-pm fraction of
kaolinite and its extraction to remove oxide and hydroxide
coatings. Both Na+- and Ca2+-saturatedkaolinites were
prepared. Electrolytes used in the experiments encomPassed high and low I (0.1 or 0.005), pH 4.5 or 6.5, and
mono- Or divalent cations LNaC104 or Ca(C104)2]. Immediately before use, the kaolinite was treated with H202
to remove oxidizable organic material (46). Hematite was
purchased from J. T. Baker Chemical Co. Trace organic
contaminants were removed by heating the hematite at
500 “covernight prior to the sorption experiments. The
BET surface areas determined by triple-point Nz(g)
adsorption were 12.5 and 5.4 m2/g for kaolinite and
hematite, respectively (Table 1).
Peat humic acid (PHA), a well-characterized terrestrial
humic substance from the International Humic Substances
0 1994 Amerlcan Chemical Society
This article is a U.S. government work, and is not subject to copyright in the United States.
Environ. Sci. Technoi., Voi. 28, No. 7, 1994 1291
~~~
~~
Table 1. Properties of Mineral Sorbents, Peat Humic Acid,
and Hydrophobic Organic Compounds
Properties of Mineral Sorbents
kaolinite
surface area (m2/g)a
sites (mol of OH-/g)b
expt surface area (m2/L)c
PHZPC
hematite
12.5
1.56 X 106
1600
7.3d
5.41
8.98 X 10-5
400
a.5=
Properties of Peat Humic Acid/
atomic aliphatic carboxyl
0
(110-165) (0-90) (165-190)
C
H
47
24
20
N
Ash
56.82 4.06 34.91 3.74 1.92
Properties of Hydrophobic Compounds
solubility
MW
(mg/L)
carbazole
log
KO,
3.29
I
H
dibenzothiophene
anthracene
m
4.38
184*3
178.2
0.075
4.45
a Surface area was measured by triple-pointNdg) adsorption and
calculated using BET equation. Kaolinite (45, 47); hematite (48,
49). c Sorbent concentration used in the HOC experiments. Apparent p H , , of the kaolinite edge (50,51).e Refs 48 and 49. f The
distribution in percent of carbon functional groups was determined
by liquid-state
NMR (ppm range shown in parentheses) (52).
Elemental analyses (ppm)were performed by Huffman Laboratories,
and oxygen was measured directly (R. L. Malcolm, personal communication, 1990, U.S. Geological Survey, Golden, CO).
Society (IHSS), was used as a representative humic
substance that would be associated with soils and sediments (Table 1). Buried peat deposits are often found in
deep formations and may provide a source of humic
substances to groundwater (53, 54). However, more
importantly, PHA is representative of the type of humic
substances likely to be associated with the solid phase
rather than the aqueous phase in groundwater systems.
Uniformly 14C-ring-labeledcarbazole (Pathfinders Laboratory Inc.), dibenzothiophene, and anthracene (Sigma
Chemical Co.) were dissolved in HPLC-grade acetone and
stored in light-excluding containers at 4 "C. The integrity
of the labeled compounds was verified by HPLC. The
radiochemical purity of carbazole was determined to be
greater than 99% with a specific activity of 9.2 Ci/mol.
Dibenzothiophene had a purity greater than 98% with a
specific activity of 1.1 Ci/mol, and anthracene had a purity
greater than 99% with a specific activity of 20 Ci/mol. A
solution of unlabeled anthracene (Aldrich Gold Label) was
also prepared using HPLC-grade acetone. The physical
properties of these compounds are shown in Table 1.
High-Performance Size Exclusion Chromatography. High-performance size exclusion chromatography
(HPSEC) was used to determine the relative molecular
sizes of PHA in solutions with different pH, I, and cation
valence. The HPSEC system consisted of a Waters 6000A
pump equipped with a Rheodyne 7125 manual injection
valve, a 1-mL injection loop, a Bio-Si1SEC 250 preparative
column (21.5 X 600 mm) (Bio-Rad Labs, Richmond, CAI,
1292 Environ. Sci. Technol., Vol. 28, No. 7. 1094
a SpectroMonitor 3100 UV-vis detector (absorbance =
254 nm) (Rainin Instruments, Woburn, MA), an H P 1046A
programmable fluorescence detector (E, = 230 nm, E, =
401 nm), and an H P 3396 Series I1 integrator (Hewlett
Packard, Corvallis, OR). The samples were run at ambient
temperature at a flow rate of 6 mL/min in either sodium
acetate or calcium acetate eluent. Aliquots of a concentrated solution of PHA, dissolved in distilled water, were
adjusted in either sodium or calcium acetate at I ranging
from 0.1 to 0.001 and at pH 4.5 or 6.5 prior to injection.
Therefore, the pH and I of PHA sample injection volume
matched those of the eluent.
Sorption Experiments. The kaolinite and hematite
were coated with PHA for the HOC sorption experiments
by equilibrating the sorbents for 20 h with varying amounts
of dissolved PHA (0--90 mg/L) in NaC104 (0.1 or 0.005
mol/L at pH 4.5) or in Ca(C104)~(0.035 or 0.0017 mol/L
at pH 4.5 or 6.5). A time-course study showed that the
sorption equilibrium of humic acid with hematite occurs
within the first hour and within 8-10 h on kaolinite. The
pH was maintained with a titrator/pH stat during the
equilibration. The suspension was then centrifuged (3000g
for 20 min), and the humic acid in the supernatant was
analyzed by measurement of the dissolved organic carbon
(DOC). The coated sorbent was washed once with DOCfree electrolyte and centrifuged, after which the DOC of
the supernatant was analyzed. The suspension was then
concentrated -4 times by decanting the supernatant and
resuspending the sorbent in a smaller volume of DOCfree electrolyte. Aliquots of the sorbent suspension were
transferred to tared, 8.5-mL, glass centrifuge tubes to
measure HOC sorption (described below). The mass
fraction organic carbon (fW)contributed by the sorbed
humic substances ranged from -0.01 to 0.53' 6 (calculated
by difference accounting for the mass of DOC at each step
of the preparation).
The radiolabeled carbazole, dibenzothiophene, or anthracene solutions were added to the concentrated sorbent
suspensions to reach a concentration of 25% of the
compound's water solubility. Large sorbent concentrations were required for statistically valid sorption measurements. The surface area concentrations of kaolinite
and hematite were 1600 and 400 m2/L (128 and 74 g/L),
respectively. Headspace in the 8.5-cm3 centrifuge tube
was 0.5 cm3 to minimize HOC volatilization, and the tube
caps were lined with aluminum foil. Light was excluded
from the anthracene experiments to prevent photodegradation. The suspensions were equilibrated with slow
end-over-end mixing for 20 h at 25 "C and centrifuged at
3000g; then 1-mL aliquots of the supernatant were
transferred to tared vials containing 12 mL of scintillation
cocktail. Each experiment was repeated with a mineral
blank containing the sorbent with no humic coating and
a "no-HOC" blank containing only the coated mineral to
determine the extent of desorption of the bound humic
substances. The pH of the supernatant was determined,
and the no-HOC blank was analyzed for DOC. Desorption
of the humic coating was insignificant (i.e., DOC was not
measurable) over the course of these experiments.
The sorbed fraction of HOC was calculated by difference
based on the 14C counts remaining in solution. In a
previous study (2), the experimental procedures were
tested by desorbing the 1%-labeled HOC and counting
the bound fraction. An average of 92 5% of the sorbed HOC
was recovered, which showed good agreement with the
values for fractional HOC sorption calculated by difference.
DibenzothiopheneSorption Isotherms at Constant
foe. Isotherms for dibenzothiophene were determined for
kaolinite (800 m2/L) with a PHA-fWof 0.05% at pH 6.5
and0.08 % at pH 4.5. Four-point isotherms were measured
(in triplicate) over the concentration range of 0.22-0.80
mg/L at pH 4.5 and 6.5 in low I(0.005) Ca(C104)zor NaC104
electrolytes. Equilibration and sampling were performed
as described above.
Competition Experiments. Kaolinite was transferred
to tared 25-mL Corex tubes to yield 800 m2/L, cleaned
(HzOz), and rinsed three times in 0.0017 mol/L Ca(C104)~
at pH 6.0. The kaolinite was then coated with PHA (foe
= 0.05 % ) as described above and rinsed two times in DOCfree 0.0017 mol/L Ca(C104)2electrolyte. Six stock solutions
were prepared containing 220 nmol/L 14C-labeledcarbazole
and variable amounts of unlabeled anthracene, ranging
from 0 to 280 nmol/L. Individual HOC stock solutions
were added to duplicate PHA-coated kaolinite samples
and an organic-free kaolinite blank. A no-HOC blank
containing the coated mineral only was used to determine
the extent of desorption of the bound PHA.
The suspensions were equilibrated and sampled as
described previously. The supernatants were analyzed
for pH, DOC, carbazole (by scintillation counting and highperformance liquid chromatography, HPLC), and anthracene (by HPLC). A C-18 (ODS) column was used
with a Waters Associates WISP 710B HPLC. The mobile
phase was 85:15 (v:v) acetonitri1e:water. An H P 1046A
fluorescence detector was used to detect the carbazole and
anthracene with an excitation wavelength of 237 nm and
an emission wavelength of 401 nm. The retention times
for carbazole and anthracene were 4.8 and 7.7 min,
respectively. The fractions of HOC sorbed were determined by difference,as described above. In the case where
the carbazole concentration was determined by the two
analytical methods (Le., 14C counts and HPLC), both
methods gave similar values.
Results and Discussion
In this paper, hydrophobic sorption is used as a general
term which encompasses two primary mechanisms: phase
partitioning and hydrophobic adsorption. Phase partitioning is the transfer of a solute from the aqueous phase
to an organic phase. The transferred solute is “solvated”
by the organic material, which surrounds it in a threedimensional orientation. Hydrophobic adsorption (also
referred to as “specific hydrophobic adsorption” in the
literature) is a surface reaction that occurs in competition
with water on a region of the surface (either organic or
inorganic) that exhibits low hydration energy. The free
energy of the sorbed complex is controlled by one- or twodimensional contact with the surface. Both of these
mechanisms are believed to be environmentally significant
for different types of materials. We note that ambiguity
exists with respect to this terminology, and controversy
does exist regarding the mechanisms of “hydrophobic“
sorption on different environmental substrates.
Configuration of Humic Substances in Solution.
The effect of solution chemistry on the molecular size of
humic substances has been determined by viscosity and
surface pressure measurements (34, 55), sedimentation
and diffusion measurements (36, 56), and size exclusion
chromatography (13,35,57,58). Although each method
has limitations, all of these studies have suggested that
humic macromolecules behave like polyelectrolytes and
that the configuration of the macromolecule in solution
will influence its structure when bound to a solid surface
(30-33).
Solute retention in HPSEC depends on differential
migration of molecules between the mobile phase and the
pore space on the stationary phase (an entropy-driven
process). Recent HPSEC studies of humic substances have
used surface-modified porous silica as a column packing
(13, 35, 57, 58) which minimizes enthalpy-driven or
nonexclusion processes (charge repulsion and adsorption).
Evidence that enthalpy-driven processes are minimal is
based on whether the eluted peaks fall between V , (void
volume) and V, (total permeation volume). The largest
molecules are excluded from the pores and elute as
nonretained molecules. The amount of mobile phase that
is required for a nonretained molecule to elute is the V,,
(defined usingdextran, MW = 162 000). Smaller molecules
have longer retention times. The volume of mobile phase
that is required for the smallest molecules to elute is called
the V, (defined using acetone). Regardless of the electrolyte, the PHA peaks were never out of the working
range of the column (before V, or after V,), indicating
that peak shift under different chemical regimes was not
significantly affected by nonexclusion phenomena ( 13,59).
Molecular weights were not calculated from the chromatograms because humic substances have a multimodal
distribution that is dependent on I , pH, and cation valence.
The HPSEC results (Figure 1)were consistent with that
reported for polyelectrolytes (30-33,37,42,43), At a given
pH, the relative size of PHA increases with decreasing I
(denoted by the shift in PHAmasstoward shorter retention
times). Others have observed this phenomenon with humic
substances using HPSEC (13,35,59)and have evoked the
random coil model (34, 56). At low I or high pH, the
electrostatic repulsion between adjacent functional groups
(e.g.,carboxyl, hydroxyl groups) is maximized, expanding
the structure of PHA and increasing its apparent molecular
size. Comparison of the chromatograms in the presence
of the sodium electrolyte (Figure la,b; pH 6.5 and 4.5,
respectively) illustrates a shift to smaller relative sizes
(denoted by longer retention times) at low pH. The
smallest, or perhaps most coiled, configuration occurred
at pH 4.5 and I = 0.1 Na+.
In calcium electrolyte (Figure lc,d), the peaks also
shifted to smaller relative sizes at higher I. However, the
elution peaks were more coalesced in the calcium electrolyte than in sodium. Bridging between ionized sites by
Ca2+may aggregate components within PHA so that they
elute as combined peaks that would otherwise show up as
distinct peaks in the sodium electrolyte. These aggregate
peaks have slightly longer retention times than their
sodium electrolyte counterparts, indicating smaller apparent molecular sizes (or greater coiling).
PHA Sorption to Mineral Surfaces. The different
combinations of pH, cation, and I used in these experiments along with their possible effects on the structure of
PHA are shown in Figure 2. This conceptual diagram was
compiled from HPSEC information on the relative size of
PHA in solution; the sorbed humic structures are speculative, as direct observations do not exist. The structure of
the sorbed humic substance is controlled (1) by the
configuration of the humic substance in solution (32,33,
601, (2) by the number of attachment points and degree
Environ. Sci. Technol., Vol. 28. No. 7, 1994
1293
Humic
Substance
in Solution
-
pH 4.5
Low I
Na
pH 6.5
Low I
HiEh I
Na Ca
Ca
&,+
I
J.
,
I
I
I
I
LigandLigandExchange? Exchange? Surface
J.
J.
J.
CoTlex?
I
I
LigandExchange?
J.
Humic
Substance
Sorbedto \rrhvu h/'%
Surface y
\\\k
\\
AI- or Fe-Oxide Surface
a @-
I = 0.005
.pH 4.5
HPSEC in Sodium
Acetate Buffer
I\
d) Ca"pH6.5
I*.,
"
"tesl
10
15
20
1=0.0051
DH 6.5
10
D
1 , 0 1 v1. 1 m 2 Minut1 a U I a O ( I "*I 0
0
,
,
!"
~
,
*
m
U
Minum
a
~
l
a
4
,
"I
Flgure 1. Highperformance slze exclusion chromatography of PHA
( 5 0 - p sample
~
size)underdlffermtpH, I. and solvatingc
a
mconditions.
(a) Na+. pH 4 . 5 (b) Na+. pH 6.5 (c) Ca2+. pH 4.5 (d) Caw. pH 6.5.
V. is the voM volume. and V, is the total permeatlon volume.
of ionization of carboxylate and hydroxyl sites on the humic
substance, and (3) by the density, distribution, and
reactivity (i.e., pKJ of hydroxyl sites and other microtopographic features of the surface. Evidence suggests
that the aqueous configuration of the humic substances
may be a primary determinant of its surface structure
(3C-33). ThePHA was adsorbed to kaolinite and hematite
under the pH and electrolyte conditions shown in Figure
2, with one exception. To enhance the sorption of PHA
to kaolinite in Na+ electrolyte for the experiments
conducted at pH 6.5, it was necessary first to adsorb the
PHA to kaolinite at pH 4.5 and then raise the pH to 6.5.
A t low pH (4.5). ligand exchange will be the primary
binding mechanism of humic substances to hydroxylated
mineral surfaces (i.e., iron and aluminum oxides; layer
silicate edges) (14, 16-18, 61). Evidence for organic
carboxylate ligand exchange is indirect, often following
the lines of evidence established for inorganic oxyanion
ligand exchange (47). Sorption of humic substances by
ligand exchange is believed to occur in three steps: (1)
1294
Envlnm. Sei. T e n d . , Vd. 28. No. 7. 1994
a
15
20
25
10
35
40
3n
35
40
\
25
Figum2. Comblnatknsof pH. I.andcationsused Intheseexprimants
and their effectson the confumatlonof PHA in solution based on size
exclusion chromatography and the hypothetlcalconfortnationof PHA
adswbed to an ox- surface.
protonation of the surfacehydroxyl (SOH + H+ = SOH*+),
rendering it more exchangeable (47,62); (2) outer-sphere
complexationof the carboxylategroup with the protonated
hydroxyl group; and (3) ligand exchange (OH? for
Hu-COO-) to yield an inner-sphere complex. The release
of hydroxylion during the adsorptionof humicsubstances
to hematite in the sodium electrolyte (pH = 4.5) has been
taken asdmct evidencefor theligandexchangemechanism
(961) and has been previously measured in our laboratory
(3).
A t high pH (6.5) and in the presence of di- and trivalent
cations, a ternary surface complex may form between the
hydroxyl site, the electrolyte cation, and ionized groups
in the humic substance (47,60,63). Among these ternary
surface complexes, a distinction can be made between
solvated cations (water-bridging) and unsolvated cations
(cation-bridging)(47,64,65). Water-bridgingimplies that
the cation remains solvated,as would occur with Ca*+and
other hard Lewis acids (47, 60). In the presence of
monovalent cations, the solvating water molecule may be
displaced yieldingan inner-spherecomplex. In this study,
kaolinitewas theonly sorbent usedwithcalcium electrolyte
Humic Substance in Solution
@mol PHA C/ml)
Humic Substance in Solution
( p o l PHA C/ml)
Cumulative pl of Titrant
(1M HCI)
Flgure 3. (a) Sorption isotherms of PHA on kaolinite in NaC104 electrolyte at I = 0.1 (0)and 0.005 (0)at pH 4.5 and in Ca(Cl0,h electrolyte
at I = 0.005 (A),pH 6.5. Dashed lines define approximate sorption maxima. (b) Sorption isotherms of PHA on hematite at pH 4.5 in NaClO,
electrolyte at I = 0.1 (0)and 0.005 (0).(c) Cumulative hydroxyl release from sorptlon isotherms of PHA on hematite (shown in b) as measured
by titrant addition to maintain constant pH at I = 0.1 (0)and 0.005 (0).
at pH 6.5. A ternary surface complex is plausible for this
system at pH 6.5 (Figure 2) because a measurable cationexchange capacity (CEC)exists for kaolinite (0.0973mol/g
at pH 6.5 for KGa-1).
The adsorption of PHA to kaolinite followed a Langmuir-type isotherm (Figure 3a) similar to those reported
by others (2,16,61). The plateau or adsorption maximum
of the Langmuir isotherm can be used to estimate the
monolayer coverage of the solid by the polymer (2, 14).
However, monolayer formation does not necessarily imply
saturation of surface hydroxyls. Langmuir isotherms of
PHA on kaolinite in Na+ electrolyte varied with I (Figure
3a). At I = 0.005, the sorption maximum of PHA occurred
at approximately 6.5 pmol of C/m2 as opposed to almost
13 pmol of C/m2 a t I = 0.1. On a mole-carbon basis, this
implied that the PHA occupied approximatelytwice (2.31
x lO-l9 m2) the surface area at low I as compared to high
I. The adsorptive maxima were consistent with the relative
sizes of PHA in solution (Figure 2; 34,351. At low I, the
adsorption maximum in the Ca2+electrolyte was similar
to that in the Na+ electrolyte (Figure 3a). The HPSEC
chromatograms for PHA under these electrolyte conditions
were also similar (Figure 11, suggesting that the PHA was
constrained to the same size range.
The mass of PHA sorbed to hematite was -1 order of
magnitude higher than that sorbed to kaolinite (Figure
3b) reflecting differences in (1) the distribution, concentration, and electrostatic environment of hydroxyl sites
on the two sorbents and (2) the intrinsic selectivity of
coordinatingmetal ion centers for carboxylate groups. Ionic
strength effects were not observed until a sorption density
of approximately 150 pmol of PHA C/m2of hematite was
reached. At high concentrations, PHA sorption displayed
non-Langmuirian behavior possibly resulting from multiple-layer adsorption or lateral interactions. At low I,
hydroxyl release was greatest at low sorption densities
(Figure 3c) suggesting that in the elongated PHA configuration there were more attachment points between
the PHA carboxyls and the iron-hydroxyls. In contrast,
at high I, hydroxyl release was less at low sorption densities,
consistent with a coiled structure (less exposure of
carboxyls on the PHA) and fewer attachment points.
Sorption of HOC to Humic-Coated Minerals in Na
Electrolyte. Previous studies of HOC binding to humatecoated sorbents indicated that sorption was more con-
sistent with an adsorption process rather than a phase
partitioning process (2). Given these observations and
the results of the isotherm experiments in the previous
section, we speculatedthat HOC sorption would be greatest
at low I where a maximum amount of humic surface area
(34-361, and therefore hydrophobic sites, on the PHA
would be exposed.
Experimental data with the Na electrolyte seemed to
confirm this hypothesis; HOC sorption was greater at I =
0.005 than at I = 0.1 (Figure 4). This trend is opposite to
a salting-out effect. The effect of I was greater for PHA
bound to kaolinite than for PHA bound to hematite.
However, and consistent with previous work ( 2 ) , the
sorption of HOC was greater on PHA associated with
hematite than on that associated with kaolinite. This
effect of I contrasts with that expected for partitioning,
where increasing I reduces the aqueous solubility of HOC
and increases partitioning to organic phases (66,67).
PHA should display its most linear, elongated structure
a t high pH and low I (Figure 1 and refs 34-36) as a result
of charge repulsion between ionized functional groups. If
the solution configuration is a determinant of surface
structure and hydrophobic adsorption is the dominant
process, then HOC sorption to the mineral-bound PHA
should be maximized under these conditions. Kaolinite
was used to test this hypothesis. Because PHA sorption
was low at pH 6.5, PHA was bound to the kaolinite at pH
4.5 to achieve the requisite adsorption density. After
coating, the pH was raised to 6.5 for the HOC sorption
experiment. Similar to the results at pH 4.5, the sorption
of dibenzothiophenewas enhanced at lowlat pH 6.5 (inset
of Figure 4). However, there was no significant difference
in the sorption of dibenzothiopheneat pH 4.5 and pH 6.5.
Apparently, the surface structure of the humic substance,
as it affects dibenzothiophene sorption, was determined
by the pH a t which the PHA was bound.
Comparative Effects of Ca2+ Electrolyte. In the
calcium electrolyte, the mineral-bound PHA may adopt
a condensed structure that is promoted by intrasite cationbridging resembling a hydrophobic phase (rather than a
hydrophobic surface). If this is the case,then the likelihood
of a phase-partitioning mechanism should be maximized
in the presence of the calcium electrolyte. Conversely, it
may be hypothesizedthat calcium facilitates the exposure
of hydrophobic binding sites to the solution phase by
Environ. Scl. Technol., Vol. 28, No. 7, 1994 1295
Z = 0.005
0
Anthracene
z = 0.1
Kaolinite Inset:
dibenzothiophene sorption
at Z = 0.005; pH 4.5 (A) and
6.5 (r)
Hematite
1
0
2
3
4
PHA Sorbed (mg C)
Kaolinite
I
0
2
PHA Sorbed (mg C)
1
0
0.5
1
1.5
PHA Sorbed (mg C)
Flgure 4. Comparison of the fractional sorption of HOC on kaolinite and hematite coated with PHA in NaCIO, electrolyte at I = 0.1 and 0.005
at pH 4.5. Kaolinite inset of dibenzothiophene sorption at constant I but varying pH.
-1.5
-1.5
.g Y -2
-2
z
%* a
0
%
.
U
as
ij
p
sw
-2.5
-2.5
-4
-3.5
-3
-2.5
log Dibenzothiophene Aqueous
(ClmoVd)
-4
-3.5
-3
log Dibenzothiophene Aqueous
-1 5
( W O Y d )
Figure 5. Sorption isotherms of dibenzothiophene on kaolinite coated with PHA in Ca2+ or Na+ electrolyte at pH 4.5 and 6.5; I = 0.005. The
inset tables show the Freundlich isotherm constants.
bridging (18,37,42,43) or neutralizing ionized sites on the
PHA (18,37,40,41). The existence of a phase-partitioning
mechanism for HOC sorption would be supported by HOC
sorption isotherms exhibiting unit slope and by a lack of
competition between like sorbates.
Sorption isotherms for dibenzothiophene on kaolinite
coated with PHA were linear on a log-log basis (Figure 5)
and conformed to the Freundlich equation:
log s = log KF -k N log c,
(1)
where KF and N are constants specific to the coated
sorbent, S equals micromoles of HOC adsorbed per gram
of solid, and C, equals micromoles of HOC per milliliter
of solution. The log-log isotherms did not exhibit unit
slope (Le., Freundlich N # 1)regardless of electrolyte. A
previous evaluation (2) showed that nonunit slopes were
observed regardless of whether sorption was measured by
aqueous-phase difference or by direct analysis of the
sorbent phase. The isotherm slopes were farthest from
unity at lower pH. Hydrophobic adsorption may yield a
linear (i.e., N = 1) or a nonlinear (Le., N f 1)isotherm;
however, phase partitioning generally yields a linear
1298 Environ. Scl. Technol., Vol. 28, No. 7, 1994
isotherm (24,66,68). The slopes of the dibenzothiophene
isotherms on the kaolinite/PHA mixture suggested that
the sorbed humic substance behaved as a hydrophobic
surface toward the HOC, characterized by hydrophobic
adsorption.
The HPSEC analyses (i.e., coalesced peaks in Figure 1)
suggested that the disordered configuration of the humic
macromolecule may be constrained in the presence of
calcium. Unlike the Na+ electrolyte, I had little effect on
dibenzothiophene sorption in the calcium electrolyte at
comparable PHA carbon concentrations (Figure 6). Apparently, dibenzothiophene sorption was insensitive to
shifts in the relative size of PHA over this range of I in
the Ca2+ electrolyte (Figure 1). We suggest that the
configuration of PHA in the presence of Ca2+may differ
from that of PHA in high I Na+. The HPSEC chromatograms in Figure la,c showed that the PHA peaks in Ca2+
and Na+ at I = 0.1 coalesce into one major peak, but that
the relative size of PHA in Ca2+is smaller (indicated by
a longer retention time). However, the sorption of
dibenzothiophene is enhanced in Ca2+(Figure 71, and given
the HPSEC results, the enhanced sorption cannot be
l
o
0
i
0 z=o.o05
0 Z=O.l
0
A Z = 0.005 pH 6.5
0
I
I n u n b I
1 , , ' , 1 " " 1 " " , " " 1 " "
0
0.5
1
1.5
2
2.5
Peat Humic Acid Sorbed (mg C)
3
Figure 6. Fractional sorption of dibenzothiophene on kaolinite and
hematite coated with PHA In Ca(C104)2electrolyte at pH 4.5 and I =
0.1 and 0.005. The sorption of dibenzothiophene on humic-coated
kaolinite at pH 6.5 and I = 0.005 is shown for comparison.
60
,
where S m i n is the sorbed HOC concentration (mol/g)
attributed to the mineral surface, &(s) is the mineral
surface-binding constant (mL/m2) as calculated from a
sorption experiment with clean mineral surface, C is the
HOC concentration in solution (mol/mL), and .3 is the
exposed surface area of the mineral (m2/g). Sorption of
the HOC was very low on the bare mineral surfaces because
they had been pretreated to remove any small amounts
of organic carbon. Sorption was nonetheless measurable
because the mineral surface areas used in the sorption
experiments were large (1600 m2/L for kaolinite and 400
m2/L for hematite). At an initial sorbate concentration
of 25% water solubility, the percent sorption of carbazole,
dibenzothiophene, and anthracene was 3, 6, and 8%,
respectively, for hematite and 6,19, and 25 % ,respectively,
for kaolinite.
Using S m i n from eq 2, a value for the organic-only
contribution to S, Sorg,
can then be calculated:
(3)
1
1
0
requires that the surface area of the mineral sorbent
exposed to the HOC be known (69);that is
NaClO
0.5
1
1.5
2
Peat Humic Acid Adsorbed (mg C)
The exposed surface area of the mineral, 4,was calculated
by assuming that the plateau in the adsorption density in
the humic/mineral isotherms (e.g., Figure 3) corresponded
to monolayer coverage. Based on this assumption, the
surface area occupied by 1mol of humic acid carbon was
calculated, and this parameter was used to determine the
fractional coverage of the mineral surface by PHA at lower
adsorption densities. The complexation of humic substances to kaolinite occurs at the edge where Lewis acid
aluminol groups are exposed. We surmise, but cannot
substantiate, that the gibbsite basal plane on kaolinite
with its doubly coordinated hydroxyls would be uncovered
by the humic substance. Lockhart (70)similarly concluded
that the basal plane of kaolinite was free of organic
substances in a humate-clay association with 6 % organic
carbon.
Single-point sorption constants (&) were calculated
according to the relation
Figure7. Comparison of dibenzothiophenesorption on kaolinite coated
with PHA in high I(O.1)Ca2+ or Na+ electrolyte at pH 4.5.
(4)
attributed to differences in PHA size under the two
different cations. Therefore, the calcium may be facilitating the exposure of hydrophobic regions on the PHA,
enhancing HOC sorption.
If hydrophobic adsorption is important, then in accordance with mass action laws, HOC with different KO,
should display competitive adsorption behavior as site
saturation is approached. By analogy, the absence of
competition is often taken as evidence of a partitioning
mechanism (7,24). Multisorbate experiments were performed by holding the carbazole concentration constant
and varying the anthracene concentration over a range in
molar ratios from 0 to over 1. Because anthracene is more
strongly sorbed than carbazole, it should suppress carbazole sorption if the sorbates indeed compete for a
common set of surface sites on the mineral-bound PHA.
The HOC single-point Kd'S for the sorbed humic phase
were calculated according to the method of Murphy et al.
(2). Calculation of the mineral contribution to sorption
where S is total adsorbed HOC (in mol/g), S m i n is HOC
adsorbed to the mineral surface in the absence of the humic
substance, and Cis the equilibrium concentration (in mol/
mL) .
The competition experiment was performed in low ionic
strength (I = 0.005) calcium electrolyte to facilitate the
potential formation of hydrophobic domains with the PHA.
Chiou et al. (7)suggested that the inability of polyacrylic
acid to enhance HOC solubility may relate to the frequent
and orderly attachment of carboxyl groups that prevent
the formation of a sizablehydrophobic domain, a constraint
that was not observed with the natural humic substances
used in their experiments. The calcium concentration in
our experiments was low enough to avoid flocculation of
the PHA, but high enough to facilitate sorption of the
PHA to the mineral surface at the higher pH (presumably
by forming a ternary surface complex). A suppression in
the Kd of carbazole in the presence of anthracene (Figure
8) was indeed noted. The carbazole Kd decreased as it
Environ. Scl. Technol., Vol. 28, No. 7, 1994
1297
M" 4 1
I
4
21
The strength of the calcium-carboxylate interaction
negates apparent I effects on HOC sorption. Changes in
Iranging from 0.1to 0.005had little effect on HOC sorption
in the calcium electrolyte. It is possible that, over the
range of I used in these experiments, sufficient calcium
ion is present to set the structure of the PHA in solution
that is preserved when the PHA is adsorbed to the mineral
surface.
In these experiments, HOC sorption was most consistent
with a hydrophobic adsorption mechanism. HOC sorption
isotherms on humic-coated minerals were nonlinear
regardless of cation, pH, or I. Carbazole and anthracene
displayed competitive sorption behavior on PHA-coated
kaolinite in low I calcium electrolyte, further supporting
the suggestion that hydrophobic adsorption rather than
phase partitioning is occurring on these natural coatings.
0
0
100
200
Anthracene Concentration
(nmoVL)
300
Figure 8. Sorptiondistributioncoefficient,Kd,of carbazole with varying
concentrationsof anthracene on kaolinlte coated with PHA at pH 6.0
in low I(0.005)Ca2+ electrolyte.
would if the carbazole was competing with anthracene for
hydrophobic adsorption sites on the mineral-bound PHA.
In a system where partitioning is occurring and the amount
of carbon sorbent is not limiting, carbazole sorption should
not be affected by a changing concentration in anthracene
(7, 24). On a mole-carbon basis, the concentration of
sorbent (PHA) was 100 times greater than the highest
combined concentration of carbazole and anthracene. An
analysis of variance showed that an average error for the
triplicate carbazole & values was h0.2,yielding narrow
error bars corresponding to the width of the symbol boxes.
Thus, the reduction in the Kd of carbazole was statistically
significant.
Conclusions
Ionic strength, pH, and electrolyte cation valence were
varied to maximize structural differences in the dissolved
humic acid, which would also influence the structure of
the sorbed humic coating. On a micromole-carbon basis,
PHA occupies approximately twice as much surface area
on kaolinite when adsorbed in sodium electrolyte at low
I than at high I. In sodium electrolyte, the sorption of
HOC to humic-coated minerals is enhanced at low I,
suggesting that the linear or open configuration of the
humic acid at low I facilitated the exposure of hydrophobic
adsorption sites.
In both calcium and sodium electrolytes,the distribution
of hydroxyl groups on the solid surfaces apparently affects
the structure of the sorbed humic acid in ways that are
important to HOC sorption. Hematite coated with PHA
always showed greater HOC sorption enhancement than
a comparable amount of PHA coated with kaolinite,
regardless of the cation or I. The hydroxyl groups on
kaolinite exist on the crystallite edge,while active hydroxyl
sites on hematite exist on all exposed crystallographicfaces.
These differences in the distribution of hydroxyl sites
apparently affect the sorbed configuration of the PHA
and HOC sorption. The hematite surface may promote
a more open or elongated structure to the PHA that, like
the low I Na+ electrolyte, is favorable to HOC sorption.
1298 Environ. Sci. Technol., Vol. 28,
No. 7, 1994
Acknowledgments
Detailed review of the manuscript by an anonymous
reviewer improved clarity and was greatly appreciated.
This research was supported by the Subsurface Science
Program, Office of Health and Environmental Research,
U.S. Department of Energy (DOE). The continued
support of Dr. F. J. Wobber is appreciated. Pacific
Northwest Laboratory is operated for DOE by Battelle
Memorial Institute under Contract DE-AC06-76RLO1830.
Literature Cited
Keoleian, G. A.; Curl, R. L. In Aquatic Humic Substances:
Influence on Fate and Treatment of Pollutants; Suffet, I.
H., MacCarthy, P.,Eds.; Advances in Chemistry Series 219;
American Chemical Society: Washington,DC, 1989;pp 231250.
Murphy, E. M.; Zachara, J. M.; Smith, S. C. Environ. Sci.
Technol. 1990,24,1507-1516.
Murphy, E.M.; Zachara, J. M.; Smith, S. C.; Phillips, J. L.
Sci. Total Environ. 1992,413-423.
Grathwohl, P.Environ. Sci. Technol. 1990,24,1687-1693.
Schlautman, M. Dissertation, California Institute of Technology, 1992.
McCarthy, J. F.;Jimenez, B. D. Enoiron.Sci. Technol. 1985,
19,1072-1076.
Chiou, C. T.; Malcolm, R. L.; Brinton, T. I.; Kile, D. E.
Environ. Sci. Technol. 1986,20,502-508.
Chiou, C. T.; Kile, D. E.; Brinton, T. I.; Malcolm, R. L.;
Leenheer, J. A.; MacCarthy, P. Environ. Sci. Technol. 1987,
21,1231-1234.
Gauthier, T. D.; Shane, E. C.; Guerin, W. F.; Seitz, W. R.;
Grant, C. L. Environ. Sci. Technol. 1986,20,1162-1166.
Gauthier, T. D.; Seitz, W. R.; Grant, C. L. Environ. Sci.
Technol. 1987,21,243-248.
McCarthy, J. F.;Zachara, J. M. Environ. Sci. Technol. 1989,
23,496-502.
Traina, S.J.; Spontak, D. A.; Logan, T. J. J.Environ. Qual.
1989,18,221-227.
Chin, Y.-P.;
Gschwend, P. M. Environ. Sci. Technol. 1992,
26,1621-1626.
Kummert, R.;Stumm, W. J. Colloid Interface Sci. 1980,
75,373-385.
Davis, J. A.; Gloor,R. Enuiron.Sci. Technol. 1981,15,12231229.
Tipping, E. Geochim. Cosmochim.Acta 1981,45,191-199.
Davis, J. A. Geochim. Cosmochim. Acta 1982,46, 23812393.
Tipping, E.; Cooke, D. Geochim. Cosmochim. Acta 1982,
46,75-80.
Tipping, E.; Heaton, M. J. Geochim.Cosmochim.Acta 1983,
47,1393-1397.
Morse, J. W. Mar. Chem. 1986,20,91-112.
(21)Inoue, K.; Zhao, L. P.; Huang, P. M. Soil Sci. SOC.Am. J.
1990,54,1166-1172.
(22) Varadachari, D.; Mondal, A. H.; Ghosh, K. Soil Sei. 1991,
91 220-227.
(23)Lobartini, J. C.; Tan, D. H. Soil Sci, SOC.Am. J. 1988,52,
125-130.
(24) Chiou, C. T.; Porter, P. E.; Schmedding, D. W. Enuiron.
Sci. Technol. 1983,17,227-231.
(25) Karickoff, S.W. J.Hydraul. Eng. 1984,110, 707-735.
(26) Garbarini, D. R.; Lion, L. W. Enuiron. Sci. Technol. 1986,
20,1263-1269.
(27) Mingelgrin,U.; Gerstl, Z. J. Enuiron. Qual. 1983,12,1-11.
(28) Ainsworth, C. C.; Zachara, J. M.; Smith, S. C. Soil Sci. Sac.
Am. J. 1989,53,1391-1404.
(29) Rutherford, D. W.; Chiou, C. T.; Kile, D. E. Enuiron. Sci.
Technol. 1992,26,336-340.
(30) Hesselink,F.Th. In Adsorption from Solutionat the Solid/
Liquid Interface; Parfitt, G. D., Rochester, C. H., Eds.;
Academic Press: London, 1983;pp 377-412.
(31) Lyklema, J. In Flocculation, Sedimentation, and Consolidation, Proceedings of the Engineering Foundation Conference, Sea Island, Georgia; Moudgil, B. M., Somasundaran, P., Eds; American Institute of Chemical Engineers:
New York, 1985;pp 3-21.
(32)Bbhmer, M. R.; Evers, 0. A.; Scheutjens, J. M. H. M.
Macromolecules 1990,23,2288-2301.
(33) Blaakmeer, J.; Bohmer, M. R.; Cohen Stuart, M. A.; Fleer,
G. J. Macromolecules 1990,23,2301-2309.
(34) Ghosh, K.; Schnitzer, M. Soil Sci. 1980,129,266-276.
(35)Tsutsuki, K.; Kuwatsuka, S. Soil Sci. Plant Nutr. 1984,30,
151-162.
(36) Cornel, P. K.;Summers, R. S.; Roberts, P. V. J. Colloid
Interface Sci. 1986,110,149-164.
(37)Foissy, A.; El Attar, A.; Lamarche, J. M. J.Colloid Interface
Sci. 1983,96,275-287.
(38)Van der Schee, H. A.; Lyklema, J. J. Phys. Chem. 1984,88,
6661-6667.
(39)Cafe, M. C.; Robb, I. D. J. Colloid Interface Sci. 1982,86,
411-421.
(40)Ramachandran, R.; Somasundaran, P. Colloids Surf. 1988,
32,319-329.
(41) Zachara, J. M.; Resch, C. T.; Smith, S. C. Geochim.
Cosmochim. Acta. 1994,58,553-566.
(42) Cleven, R. F. J. M.; de Jong, H. J.; van Leeuwen, H.P. J.
Electroanal. Chem. 1986,202,57-68.
(43) Belton, D.; Stupp, S. 1. Org. Coat. Plast. Chem. 1980,42,
221-226.
(44)Johnston, C. T.; Sposito, G.; Birge, R. R. Clays Clay Miner.
1985,33,483-490.
(45) Zachara, J. M.; Cowan, C. E.; Schmidt, R. L.; Ainsworth, C.
C.Clays Clay Miner. 1988,36,317-326.
(46) Klute, A. Methods of Soil Analysis, Part 1 Physical and
Mineralogical Methods, 2nd ed.; American Society of
Agronomy, Soil Science Society of America, Inc.: Madison,
WI, 1986;pp 95-97.
(47) (47)Sposito, G. The Surface Chemistry of Soils; Oxford
University Press; New York, 1984;pp 138-147.
(48) Yates, D.E. Ph.D. Thesis, The University of Melbourne,
Australia, 1975.
(49)Liang, L.; Morgan, J. J. Aquat. Sci. 1990,52,32-55.
(50) Ferris, A.P.; Jepson, W. B. J. Colloid Interface Sci, 1975,
61,32-43.
(51) Rand, B.; Melton, J. E. Nature 1975,257,214-216.
(52) Thorn, K.A.; Folan, D. W.; MacCarthy, P. Water-Resour.
Inuest. Rep., US.Geol. Surv. 1989,No. 89-4196,93pp,
(53) Thurman, E. M. In Humic Substances in Soil, Sediment,
and Water; Aiken, G. R., McKnight, D. M., Wershaw, R.
L., MacCarthy, P., Eds.; John Wiley and Sons: New York,
1985.
(54) Wassenaar, L.; Aravena, R.; Fritz, P.; Barker, J. Org.
Geochem. 1990,15,383-396.
(55) Chen, Y.; Schnitzer, M. Soil Sci. SOC.Am. J. 1976,40,866872.
(56) Cameron, R. S.;Thornton, B. K.; Swift, R. S.; Posner, A.
M. J. Soil Sci. 1972,23,394-408.
(57) Vartiainen, T.;Liimatainen, A.; Kauranen, P. Sci. Total
Environ. 1987,62,75-84.
(58) Miles, C. J.; Brezonik, P. L. J.Chromatogr. 1983,259,499503.
(59) Crum, R. H.M.S. Thesis, Washington State University,
Pullman, 1992.
(60) Theng, B. K. G. Formation and Properties of Clay-Polymer
Complexes; Developments in Soil Science 9; Elsevier:
Amsterdam, 1979.
(61) Parfitt, R.L.;Fraser, A. R.; Farmer, V. C. J.Soil Sci. 1977,
28,297-305.
(62) Zhang, P. C.; Sparks, D. L. Soil Sci. SOC.Am. J . 1989,53,
1028-1034.
(63)Tipping, E. In Organic Acids in Aquatic Ecosystems;
Perdue, E. M., Gjessing, E. T., Eds.; Dahlem Workshop
Reports, Life Sciences Research Report 48; John Wiley &
Sons: New York, 1990.
(64) Mortland, M. M. Adv. Agron. 1970,22,75.
(65) Greenland, D.J.; Hayes, M. H. B. The Chemistry of Soil
Processes; Wiley: Chicester, U.K., 1970.
(66) Chiou, C. T.;Peters, L. J.; Freed, V. H. Science 1979,206,
831.
(67) Chiou, C. T.; Peters, L. J.; Freed, V. H. Science 1981,213,
683-684.
(68) Chiou, C. T.; Stroup, T. D. Enuiron. Sci. Technol. 1985,19,
1196-1200.
(69) Curtis, G.P.; Reinhard, M.; Roberta, P. V. In Geochemical
Processes at Mineral Surfaces; Davis, J. A., Hayes, K. F.,
Eds.; ACS Symposium Series 323; American Chemical
Society: Washington, DC, 1986.
(70)Lockhart, N. C. Clays Clay Miner. 1981,29,413-422.
Receiued for review October 6, 1993.Revised manuscript received March 17, 1994. Accepted March 21, 1994."
Abstract published in Advance ACS Abstracts, April 15, 1994.
Environ. Sci. Technol., Vol. 28, No. 7, 1994
1299